## Reform of the Policy on Public Debt Limits in IMF-Supported Programs (ppea2020061) ## Source details **Canonical URL:** [Reform of the Policy on Public Debt Limits in IMF-Supported Programs (ppea2020061)](https://www.imf.org/-/media/files/publications/pp/2020/english/ppea2020061.pdf) ## Other formats - [Markdown version](/-/media/files/publications/pp/2020/english/ppea2020061.pdf.md) - [Structured JSON version](/-/media/files/publications/pp/2020/english/ppea2020061.pdf.json) --- ### Introduction and core reform thrust - The architecture of the DLP would evolve rather than change radically, standardizing successful existing practices. - High-level balance: more flexibility where appropriate (tailored PV limits, NCB exceptions, tailored treatment for market-integrated borrowers) combined with safeguards (higher scrutiny where borrowing plans worsen shock-absorption capacity; tighter rules on repeated NCB exceptions). - Emphasis on greater debt data disclosure to the IMF and a tighter, single concessionality definition. ### Debt data disclosure and creditor composition - Expectation: critical debt data disclosure gaps should be addressed in Fund-supported programs upfront, premised on a risk-based approach. - Disclosure to Fund staff would lead to conditionality only if the vulnerability revealed is deemed critical for achieving program goals or for monitoring program implementation; case-by-case approach. - Program documents required to include a table with a profile of the holders of the country’s public debt (most recent data; updated at each program review), with collateralized debt included as a memo item; missing elements expected to be filled in by the second review at latest. - Debt subject to non-disclosure agreements should be included in a special line item. - Publication of such data must be consistent with the Fund’s legal framework for confidential information and the Fund’s transparency policy. - Risk-based “red flags” to prioritize disclosure efforts include: - (i) state/local governments with borrowing capacity not captured in debt numbers; - (ii) large SOEs with external borrowing capacity not captured in the debt perimeter; - (iii) PPPs or SPVs not adequately evaluated or monitored; - Missing terms/conditions (e.g., collateral, margin call clauses, step-up rates, non-standard event-of-default clauses) combined with signs of vulnerability-enhancing conditions. ### Tailoring conditionality to financing mix and PV-based limits - For countries that normally rely on concessional financing but have significant market access, the default for moderate/high risk is a PC on accumulation of public and publicly-guaranteed external debt specified in present value (PV) terms. - Eligibility for PV treatment requires meeting requirements in SM/20/157 (page 29): having had significant access to international financial markets in recent years or access to these markets being a key element of the program, and demonstrating capacity to manage significant market borrowing. - “Significant access” operational criteria: - Cumulative borrowing of at least 50 percent of quota in at least two of the preceding three years; or - If the prior-borrowing criterion is not met, program envisages financing on international financial markets of at least 100 percent of quota over the program period (the 100 percent threshold applies only in this alternate case). - For non-concessional-financing-reliant countries, debt conditionality can take the form of limits on total public debt or targeted debt limits depending on vulnerabilities. ### Use of PV limits, nominal NCB limits, and capacity considerations - Broaden the use of PV limits for countries that normally rely on concessional financing and do not have significant market access and are assessed at moderate risk of debt distress. - Most members can be expected to have adequate capacity to monitor aggregate debt conditionality in PV terms. - Where capacity to monitor debt conditionality is not assessed as adequate, retain specification as a limit on non-concessional borrowing (NCB) in nominal terms plus a memo item limit on concessional borrowing in nominal terms. - Capacity assessments draw on the 3-year average CPIA Debt Policy score, recent DeMPA and PEFA ratings when available, and recent track record and technical assistance reports; assessment conducted in consultation with authorities. - Evidence: share of countries at moderate or high risk deemed to have adequate capacity to record/monitor debt declined from 43 percent (in 2016–17) to 33 percent (in 2018–19), mainly reflecting LIC-DSF linked downgrades. ### Zero NCB presumption and exceptions; signal-based approach - Presumption: retain zero NCB limit for countries that normally rely on concessional financing without significant market access and that are assessed to be at high risk of debt distress or in debt distress. - Proposed clearer circumstances for NCB exceptions include: - (i) project exceptions determined using a signal-based approach for projects integral to national development where concessional financing is not available; - (ii) debt management operations that demonstrably improve liquidity/solvency debt burden indicators without adverse effect on risk rating; - (iii) repeated NCB exceptions subject to higher scrutiny and additional signals; repeated debt-management exceptions may trigger reassessment of market-access status. - Safeguard: require an indicative target (IT) on public external borrowing specified in PV terms to inform and constrain exceptions; where capacity is weak, retain existing memo item on concessional borrowing (CB). - Box 5 signal-based approach (indicative core signals): - At least one signal that the project is integral to development (examples: part of PSIP/NDP, endorsed by MDB, PIMA performance >=2, credible independent assessment) and one signal that concessional financing is not available (examples: recent donor conference outcomes, explicit MDB indication of non-availability). - Quality of signals assessed case-by-case; guidance note will describe assessment detail. ### Concessionality definition, blended finance, and collateralized borrowing - Proposal: single standard concessionality threshold of 35 percent (existing standard), with allowance to apply a higher threshold when integral to restoring debt sustainability. - Specific proposed treatments: - Blended financing arrangements that include a financially significant amount of grants-in-kind to be treated as non-concessional (zero grant element). - Financing involving unrelated collateral (e.g., general budgetary borrowing collateralized with future commodity export revenues) to be treated as non-concessional (zero grant element). - Elimination of the “high concessionality tool” (previously allowed raising threshold above 35 percent in some cases); recorded usage has been limited (two cases since July 2015: Afghanistan and Central African Republic). - Rationale: these changes aim to prevent circumvention of debt limits and account for hidden costs and risk-reducing effects of unrelated collateral not captured in standard grant-element calculations. ### Capacity building, conditionality, and coordination - Reform proposals provide incentives to improve debt management capacity (access to flexible PV limits conditioned on adequate capacity; operating within PV limits incentivizes broader debt management improvements). - Continued use of structural conditionality is encouraged when significant weaknesses in debt management capacity are identified, consistent with the Fund’s Guidelines on Conditionality. - Timely capacity development support (including IMF technical assistance or other providers) will be needed to improve debt management capacity in some cases. - Directors underscored importance of close alignment between the Fund’s DLP and the World Bank’s Sustainable Development Finance Policy (SDFP). ### Observance of debt limits, compliance rates, and implementation record (July 2015–April 2020) - Sample: 87 IMF-supported programs ongoing between July 2015 and April 2020: 54 PRGT-supported and 33 GRA-supported programs (FCLs, RCFs/RFIs, and SMPs excluded). - Compliance rates: - Compliance rate for debt performance criteria (PCs): close to 95 percent. - Compliance rate for fiscal PCs: 80 percent. - Debt PCs breached in 8 programs with at least one review, representing approximately 4 percent of all test dates. - Selected breach examples (size and corrective actions preserved): - Guinea — Nominal non-zero NCB, Review R2, Size: 1 percent of limit / 0.1 percent of GDP; corrective action: QPC modified to exclude debt later cancelled or renegotiated to become concessional. - Cameroon — Nominal non-zero NCB, Review R1, Size: 64 percent of limit / 0.3 percent of GDP; corrective action: Limit for following year lowered by same amount. - Malawi — Zero NCB Limit, Review R1, Size: US$127 million / 1.9 percent of GDP; corrective action: TMU modified to exclude domestically denominated Treasury Notes and Bills from the QPC. - Liberia — PV Limit, Review R7, Size: 88 percent of limit / 0.4 percent of GDP; corrective action: Debt management unit began monthly reporting on all signed/ratified loans. - Mozambique — PV Limit, Review R1, Reason: Reporting — disclosure of hidden debt by SOEs revealed breach; program went off track. - Main reasons for non-observance: weak public financial management and debt monitoring, reporting slippages, technical factors, exogenous shocks; when significant non-observances occurred, program targets were revised to address causes in PRGT cases and debt vulnerabilities did not deteriorate subsequently. ### Containment of debt risks in IMF-supported programs (findings) - PRGT-supported programs: - Over 2015–2018, the increase in the median public debt-to-GDP ratio was lower for countries with PRGT-supported programs compared with countries without programs. - Over 2016–2019, the median public debt-to-GDP ratio declined for countries with PRGT-supported programs, compared with a significant increase for countries without such programs. - Debt risk ratings deteriorated in only a handful of PRGT cases (examples cited: Kenya, Mozambique, Sierra Leone; exogenous factors in Haiti and Chad). - GRA-supported programs: - Median public debt-to-GDP ratio declined in countries with GRA-supported programs, while it increased in countries without such programs. - Exceptions where public debt-to-GDP increased during program periods were driven by weaker-than-expected GDP growth, fiscal slippages, and larger-than-projected currency depreciations (examples: Angola, Argentina, Georgia, Jordan, Tunisia). - Shifts post-2014 DLP reform: - In PRGT-supported programs active July 2015−April 2020: 24 percent had debt limits in the form of PV limits; 11 percent had no debt conditionality. - In GRA-supported programs July 2015−April 2020: some 69 percent of programs had some form of debt conditionality (67 percent in January 2010-June 2015). ### Debt exposures outside fiscal/debt coverage perimeter and related risks - PPPs: usage in countries subject to NCB limits increased by an average 0.8 percent of GDP since program inception, compared with 0.4 percent of GDP for comparable pre-program periods; some programs introduced additional conditionality to limit PPPs or commitments to assess new PPPs. - SOEs and SPVs: in some programs, SOEs with large fiscal risks were not covered by debt conditionality due to lack of data; borrowing by these entities has increased (examples: Egypt, Seychelles, Tunisia). SPVs with significant borrowing may remove legal obligations while creating implicit contingent liabilities (example: Ghana). - Incomplete disclosure of borrowing terms/conditions has been problematic (examples: Angola, Ecuador); staff analysis suggests 44 countries in IMF-supported arrangements with red flags (high risk of debt distress and high potential for collateralized debt). ### Issues and reform focus for countries that normally rely on concessional financing - Heterogeneity increased: many countries rely on mixed concessional and non-concessional official financing; NCB scale has increased and some countries have begun significant market access. - Tension in 2014 framework for market-access cases: NCB largely precluded for high-risk countries but market financing can be embedded in program design; debt management exceptions have proven cumbersome and misaligned with market timing (example: Ghana). - Evidence of overly restrictive application of NCB access: - In high-risk cases, exceptions to NCB were recorded only in two of nine cases. - In five moderate-risk programs where NCB was feasible, NCB was set at zero throughout programs in three cases (Guinea-Bissau, Malawi, Togo). ### Alignment with World Bank SDFP and transitional arrangements - Close alignment intended with World Bank SDFP (coordination on debt policy commitments, borrowing limits, and exceptions). - Transitional mechanics: - Revised DLP to take effect following issuance of a staff guidance note; expected Guidance Note issuance in February 2021 and policy effectiveness in March 2021. - Review of implementation experience no later than five years after effectiveness; Board update no later than two years after effectiveness. - Conditionality in pre-existing programs to be modified in program discussions concluded after March 31, 2021; programs approaching penultimate/last reviews not required to change established framing. ### Illustrative guidance and tailoring of conditionality - Box 1 and supplementary illustrative guide summarize: - Purpose: debt conditionality included when significant debt vulnerabilities exist or when debt targets complement fiscal conditionality. - Form/calibration: PV-limits, nominal NCB limits, limits on total public debt, or targeted limits depending on vulnerabilities and capacity. - Debt coverage: normally public and publicly-guaranteed debt or targeted subcomponents; contracting basis preferred when project loans disbursed over extended periods. - Program documents required to include a borrowing plan as basis for quantitative limits and baseline for assessing non-observance causes. *Source: Reform of the Policy on Public Debt Limits in IMF-Supported Programs (ppea2020061) — excerpt. ### introduction of an explicit expectation that critical debt data disclosure gaps should be ### introduction of an explicit expectation that critical debt data disclosure gaps should be addressed in Fund-supported programs upfront, premised on a risk-based approach ### Debt data disclosure and creditor composition - Expectation introduced that critical debt data disclosure gaps should be addressed in Fund-supported programs upfront, premised on a risk-based approach. - Disclosure to Fund staff would only be expected to lead to conditionality if the vulnerability revealed by such disclosure is deemed critical for achieving the goals of the Fund-supported program or for monitoring its implementation. - Information on creditor composition can help strengthen program design and contribute to the broader goal of improving debt transparency. - Requirement supported that program documents include a table with a profile of the holders of the country’s public debt, with provision of supporting technical assistance where needed. - Many Directors called for the table to include debt service in addition to debt stock, wherever feasible. - Debt subject to non-disclosure agreements should be included in a special line item in the table. - Missing elements would be expected to be filled in, at the latest, by the time of the second review of the program. - Publication of such data must be consistent with the Fund’s legal framework for the treatment of confidential information as well as the Fund’s transparency policy. ### Tailoring conditionality to financing mix and PV-based limits - For countries that normally rely on concessional financing but have access to international financial markets on a significant scale, a tailored approach and better alignment of conditionality with the country’s financing mix and program design is needed. - Where such countries are assessed to be at moderate or high risk or in debt distress, a performance criterion on the accumulation of public and publicly-guaranteed external debt, specified in present value (PV) terms, would be the default choice. - Alternative formulations to PV limits may be used where warranted to better address critical vulnerabilities. - Eligibility for PV treatment requires meeting the requirements specified in SM/20/157 (page 29): having had significant access to international financial markets in recent years or access to these markets being a key element of the program, and also having a demonstrated capacity to manage significant levels of market borrowing. ### Use of PV limits, nominal NCB limits, and capacity considerations - Broadening the use of PV limits should be expanded for countries that normally rely on concessional financing and do not have significant access to international financial markets, and that are assessed at moderate risk of debt distress. - Most members can be expected to have adequate capacity to monitor conditionality on aggregate debt levels in a manner that allows use of debt limits specified in PV terms. - Where the member’s capacity to monitor debt conditionality on aggregate debt levels is not assessed to be adequate, the specification of debt conditionality as a limit on non-concessional borrowing (NCB) in nominal terms, and a memo item as a limit on concessional borrowing in nominal terms, should be retained. - Capacity would be assessed in consultation with authorities and, where relevant, informed by past experience on the quality of the member’s debt monitoring. - Directors supported higher scrutiny of borrowing plans for countries at moderate risk of debt distress with limited space as an additional safeguard. ### Zero NCB presumption and exceptions - Directors concurred that the presumption of a zero NCB limit should be retained for countries that normally rely on concessional financing without significant access to international financial markets and that are assessed to be at high risk of debt distress or in debt distress. - Proposals for greater clarity on circumstances under which exceptions to the zero NCB rule would be accommodated include: - (i) use of the signal-based approach for determining when a project is integral to the authorities’ national development program and for which concessional financing is not available; - (ii) debt management operations; and - (iii) repeated NCB exceptions. - Requirement, in these circumstances, to include an indicative target on public external borrowing specified in PV terms to safeguard debt sustainability. - Many Directors called for caution in granting exceptions; exceptions should be limited to projects that credibly generate good social and economic return and contribute to reducing overall debt vulnerabilities. ### Concessionality definition, blended finance, and collateralized borrowing - Directors supported proposed adjustments to the definitions of concessionality, including to help prevent circumvention of debt limits. - Blended financing arrangements that include the provision of a financially significant amount of grants-in-kind should be treated as non-concessional. - A few Directors urged staff to exclude grants-in-kind where fair value has been assessed from this treatment. - Financing involving unrelated collateralized debt—e.g., general budgetary borrowing collateralized with future commodity export revenues—should be treated as non-concessional. - Many Directors encouraged borrowers and creditors to carefully consider the hidden costs inherent in these financing arrangements. - Directors generally supported the application of a single definition of concessionality (35 percent) to all cases, agreeing that a higher concessionality threshold would still be allowed in cases when this is deemed to be an integral part of restoring debt sustainability. ### Capacity building, conditionality, and coordination - Reform proposals would provide incentives to improve debt management capacity. - Continued use of structural conditionality is encouraged when significant weaknesses in debt management capacity are identified in consultation with authorities, in a manner consistent with the Fund’s Guidelines on Conditionality. - Timely capacity development support (including through technical assistance provided by Fund staff or through other providers, where available) will be needed in some cases to improve debt management capacity. - Directors underscored the importance of close alignment between the Fund’s DLP and the World Bank’s Sustainable Development Finance Policy. ### Implementation timing, reviews, and outreach - Directors agreed that the DLP should take effect following the issuance of a staff guidance note as specified in the proposed decision, with expected effectiveness in March 2021. - A review of the experience in implementing this new policy would be conducted no later than five years after the entrance into effect of the new policy, with an update to the Board on the implementation of this policy no later than two years after the date of effectiveness. - Directors called for an effective outreach strategy to ensure that the reformed policy is clearly understood by stakeholders. - Many Directors encouraged all official creditors to engage with Paris Club and to follow responsible and transparent lending practices. - Document date: October 1, 2020 *Source: Reform of the Policy on Public Debt Limits in IMF-Supported Programs (excerpt).* ### Box 1. Key Elements of the Debt Limits Policy ### Box 1. Key Elements of the Debt Limits Policy ### Purpose and conditionality triggers - Public debt conditionality is normally included in IMF arrangements when a member faces significant debt vulnerabilities or when there are merits to using debt targets instead of, or as a complement to, "above-the-line" fiscal conditionality. - Debt conditionality is implemented: - For countries that do not normally rely on external financing on concessional terms: if significant vulnerabilities are identified in the Market-Access Countries Debt Sustainability Analysis (MAC-DSA) that are not adequately addressed by fiscal conditionality. - For countries that normally rely on external financing on concessional terms: the assessment of debt vulnerabilities is informed by the Low-Income Countries Debt-Sustainability Framework (LIC-DSF). ### Form and calibration of limits - For non-concessional-financing-reliant countries: - Debt conditionality takes the form of either limits on total public debt or targeted debt limits, depending on the nature of debt vulnerabilities. - For concessional-financing-reliant countries (LIC-DSF informed): - Low risk of external debt distress: limits on public external borrowing are not required. - Moderate risk of external debt distress: limits take the form of a PV-limit on public and publicly guaranteed external debt, provided capacity to record and monitor external debt is assessed as adequate. - High risk of external debt distress: a performance criterion (PC) is set on the nominal level of non-concessional external borrowing (NCB), with NCB precluded unless a compelling case can be made for it. A PC or indicative target (IT) on concessional external borrowing (CB) is also included. ### Exceptions and special cases - If capacity to record and monitor contracting of debt is weak, the PC should take the form of a limit on the accumulation of non-concessional external debt, supplemented by a memorandum item on the accumulation of concessional external debt in nominal terms. - In countries with significant financial integration into international financial markets, a limit on total public debt accumulation may be more appropriate than a limit on external debt. ### Debt coverage, measurement basis, and documentation - Debt limits normally cover public and publicly-guaranteed debt or targeted subcomponents of such debt, unless program objectives or institutional circumstances warrant otherwise. - Limits may be set on a contracting basis or a disbursement basis, with a preference for the contracting basis when external financing consists mostly of project loans disbursed over an extended period. - Program documents are required to include a borrowing plan that: - Serves as a basis for deriving quantitative limits. - Serves as a baseline for assessing causes of non-observance of debt conditionality and for requests for modifications to debt conditionality at program reviews. *Source: ppea2020061 - Box 1. Key Elements of the Debt Limits Policy* ### 17. While overall debt risks have trended up across both LICs and EMs, they appear to ### 17. While overall debt risks have trended up across both LICs and EMs, they appear to have been contained for those countries with IMF-supported programs: ### Containment of debt risks in IMF-supported programs - Sample considered: 87 IMF-supported programs ongoing between July 2015 and April 2020: 54 programs supported by the PRGT (“PRGT-supported programs”) and 33 countries not supported by the PRGT (“GRA-supported programs”). FCLs, RCFs/RFIs, and SMPs are not included in the analysis. - PRGT-eligible countries: - Over 2015–2018, the increase in the median public debt-to-GDP ratio was lower for countries with PRGT-supported programs compared with countries without programs. - Over 2016–2019, the median public debt-to-GDP ratio declined for countries with PRGT-supported programs, compared with a significant increase for countries without such programs. - In PRGT-supported programs, debt risk ratings deteriorated in only a handful of cases, reflecting fiscal slippages (examples: Kenya, Mozambique, and Sierra Leone) and exogenous factors (examples: a natural disaster in Haiti; liquidity pressures following commodity price and security shocks in Chad). - GRA-supported programs: - Debt vulnerabilities appear to have been reduced in most GRA-supported programs over the period of 2015–2019. - The median public debt-to-GDP ratio declined in countries with GRA-supported programs, while debt ratios in countries without such programs increased. - Exceptions where public debt-to-GDP increased during program periods were driven by weaker-than-expected GDP growth, fiscal slippages, and larger-than-projected currency depreciations (examples: Angola, Argentina, Georgia, Jordan, and Tunisia). ### Role of fiscal and debt conditionality; implementation record - Anchoring fiscal policy has been central to program success in containing debt vulnerabilities. - Strong record of adherence to debt conditionality with relatively tight implementation (few modifications and exceptions). - Compliance rates: - Compliance rate for debt PCs stands at close to 95 percent. - Compliance rate for fiscal PCs stands at 80 percent. - Reasons for non-observance of debt limits included weak public financial management and debt monitoring (Cameroon, 2017 on two occasions; Iraq, 2016); reporting slippages (Guinea, 2017; Liberia, 2015; Mozambique, 2015); or technical factors (Malawi, 2018; Mauritania, 2017). - When non-observances were significant, program targets were subsequently revised to address contributing factors or to limit space for new borrowing; in none of these PRGT-supported programs did debt vulnerabilities deteriorate. - In programs that went off-track, primary drivers for non-observance of debt limits were either exogenous shocks or weak fiscal performance (exception: Mozambique (2015), where debt governance weaknesses were key). ### Shifts in nature of debt conditionality (post-2014 DLP reform) - In PRGT-supported programs active during July 2015−April 2020: - 24 percent had debt limits in the form of PV limits. - 11 percent had no debt conditionality. - In GRA-supported programs ongoing in July 2015−April 2020: - Some 69 percent of programs had some form of debt conditionality, compared with 67 percent of programs in January 2010-June 2015. - Given the modest number of cases, it is not possible to provide econometric evidence on the impact of debt conditionality on debt vulnerabilities. ### Stakeholder feedback on the Debt Limits Policy (DLP) - Feedback from country authorities, MDBs, OECD export credit agencies, and IMF mission chiefs suggests the DLP has been helpful in containing debt vulnerabilities. - Perceived benefits: helped authorities prioritize investments, maximize borrowing on concessional terms, and better contain debt vulnerabilities by improving control over line ministries and SOE debt. - Concerns: policy constrained investment in some instances (including by denying semi-concessional loans); stakeholders suggested more flexibility to tailor debt limits to country-specific circumstances. ### Debt data disclosure gaps and related risks - Debt exposures outside fiscal/debt coverage perimeter generate risks: - PPPs: usage in countries subject to NCB limits increased by an average 0.8 percent of GDP since program inception, compared with 0.4 percent of GDP for periods of similar duration prior to the program. - In response, some IMF-supported programs introduced additional conditionality to limit PPPs (example: Burkina Faso, 2018) or MEFP commitments to assess the development of new PPPs (example: Niger, 2016). - SOEs and SPVs: in some programs, SOEs involving large fiscal risks were not covered by debt conditionality often due to lack of data; borrowing by these entities has increased (examples: Egypt, Seychelles, and Tunisia). - SPVs involving significant borrowing have been established and may remove the government from legal obligations while still involving an implicit contingent liability (example: Ghana). - Incomplete disclosure of borrowing terms and conditions has been problematic in recent cases: - Angola and Ecuador: due diligence supported by prior actions and structural benchmarks revealed collateralized or collateral-like debt structures. - Staff analysis suggests 44 countries in IMF-supported arrangements with red flags, defined as: (i) the risk of debt distress is high, and (ii) the potential for the presence of collateralized debt is assessed to be high (because use of collateral in loans or bonds is known to be high; there is substantial lending from creditors that typically demand collateral; or commodities, whose future flow receivables can serve as collateral, make up a high share of total exports). ### Issues for countries that normally rely on concessional financing - Heterogeneity within this group has increased: - Most countries rely on a mix of concessional and non-concessional borrowing from official sector lenders; the scale of NCB has increased over time in many countries. - Some countries have begun to access international financial markets on a significant scale, with borrowing plans envisaging continued market access during IMF-supported programs that may be inconsistent with NCB limits. - Tension in the 2014 policy framework for countries with significant market access: - For countries at high risk of debt distress, NCB can be accommodated only under exceptional circumstances, yet external financing from private financial markets can be embedded in program design. - The “debt management operations” exception can manage this tension in principle, but the process has proven cumbersome with potentially long lead times that do not align well with changing market conditions (e.g., due to bi-annual reviews). - Unintended consequences have included bias toward more expensive domestic financing and difficulties for debt management operations and issuances (example: Ghana). - For moderate risk countries in this subgroup, use of a limit on total nominal public debt is possible but appropriate only in very specific circumstances (noted as chosen in only two cases: Kenya and Senegal). ### Issues for countries with no significant market access - Shift to PV limits for countries at moderate risk of debt distress: - Requirements for allowing use of PV-limits appear to have been too demanding. - The methodology for assessing debt recording/monitoring capacity captures broad debt management competencies rather than the narrow capacity to record/monitor debt (the key factor for deploying quantitative conditionality on aggregate debt levels). - Evidence on capacity assessments: - Decline in share of countries at moderate or high risk of debt distress deemed to have adequate capacity to record/monitor debt: from 43 percent (in 2016–17) to 33 percent (in 2018–19). - This decline is due mainly to downgrades of countries linked to a shift to a high risk of debt distress in the LIC DSF, which feeds into CPIA debt scores; there is little evidence this downgrade reflects pure weakening in recording/monitoring capacity. - Access to NCB was tightly restricted, beyond expectations: - In IMF-supported programs for countries at high risk of debt distress, where exceptions to NCB are possible, these were recorded in only two out of nine cases (albeit repeatedly) (examples: Cameroon and Mauritania). - In five IMF-supported programs for countries at moderate risk of debt distress, where NCB is feasible, NCB was instead set at zero throughout programs in three cases (examples: Guinea-Bissau, Malawi, and Togo), with no exceptions granted. ### Measurement issues for debt recording and monitoring capacity (Box 3 summary) - Assessment draws on 3-year average CPIA Debt Policy score, and recent DeMPA and PEFA ratings when available; final assessment also considers recent track record and technical assistance reports. - Identified problems: - CPIA debt score has broad debt management coverage, including elements not strictly related to recording/monitoring capacity (e.g., contribution of debt management strategy to minimizing budgetary risks and ensuring long-term debt sustainability). - Averaging of the CPIA debt score can delay reflection of recent deteriorations or improvements in capacity. - Scoring inertia: DLP Guidance Note prescribes that upgrades of capacity should be irreversible absent exceptional circumstances, which may slow upgrading of capacity. *Source: Excerpt from IMF chapter titled “Reform of the Policy on Public Debt Limits in IMF-Supported Programs” (pages provided).* ### 28. The guidance on when to accommodate NCB exceptions for countries at high risk of ### 28. The guidance on when to accommodate NCB exceptions for countries at high risk of debt distress may have been difficult to interpret ### Ambiguity and implementation problems for NCB exceptions - NCB exceptions for projects: - Can be granted when they are considered integral to the authorities’ development program and concessional financing is not available. - The 2014 policy provides little guidance on how to assess “integral” or how to assess debt sustainability implications for projects. - The 2014 policy is silent on handling repeated requests for project exceptions. - NCB exceptions for debt management operations: - Can be granted when they “improve the overall debt profile.” - What constitutes such an “improvement” is not well-specified. - Consequence: lack of clarity has made policy implementation difficult for authorities to understand and for staff to explain. ### Contextual note from accompanying figure/text - Note: 3.0 is the capacity assessment threshold. *Source: The World Bank data referenced in the original text.* --- ### Definition of concessionality (B.2.3) — thresholds and anchors - The concessionality threshold: - 35-percent grant element threshold for determining a concessional loan (dates back to 1995). - 5 percent discount rate (informed by the 10-year average Commercial Interest Reference Rates (CIRR) rate and the per capita income growth rate). - Both CIRR and per capita income growth have declined by about 0.5 percentage points since the last review of the LIC-DSF in 2017. - Debtor country view: threshold has worked well to encourage concessional lending but may have discouraged some semi-concessional financing. ### Difficulties in evaluating concessionality - Blended financing packages: - Concessionality can be jointly assessed if combined instruments are regarded as the integrated incurrence of debt. - Challenges arise for non-conventional project financing where grant element provided in kind—difficult to evaluate due to lack of standard competitive procurement procedures or independent third-party assessment. - Staff cannot always verify whether the concessional element is undermined by over-invoicing in other dimensions of a broader deal. - Collateral: - Standard concessionality estimation uses only the nominal interest rate and does not account for the value of collateral, which may reduce lender risk and lower nominal rates. - When collateral is standard practice (e.g., trade or project finance) there may be no circumvention problem; when collateral is “unrelated,” there can be both circumvention and debt vulnerability issues. - Unrelated collateral can increase credit risks for new, uncollateralized loans and raise their cost. ### High concessionality tool effectiveness - The DLP permits increasing the concessionality threshold well above the standard 35 percent for particularly vulnerable debt situations. - Recorded usage: only two cases since July 2015—Afghanistan and Central African Republic. - Evidence is scant that the 2014 policy led to a meaningful increase in concessionality of financing; for many donors, use of grants was already pre-determined by the LIC DSF rating. --- ### Tailoring conditionality for countries that do not normally rely on concessional financing - Overall tailoring generally appropriate, with scope for improvement. - Examples of conditionality used: - Ceilings on total public debt: Barbados, Cyprus, Egypt, Jordan, Jamaica. - Ceilings on external debt: Equatorial Guinea, Gabon. - Limits capturing specific vulnerabilities: short-term debt (Jordan), collateralized debt (Angola). - Issues identified: - Overlapping conditionality with fiscal targets (Egypt) implying debt conditionality may have been redundant. - Some programs lacked limits capturing government guarantees where SOE sector posed fiscal risks (Egypt, Seychelles, Tunisia). - Recommendation: future programs must rigorously identify and target debt vulnerabilities for conditionality. --- ### Reform proposals — overall approach (REFORM PROPOSALS) - Assessment: - 2014 policy broadly worked well, but improvements needed in several areas. - For countries that normally rely on concessional financing, needs include: - Accommodating special circumstances of countries with significant access to international financial markets. - Removing unwarranted impediments to broader use of PV limits. - Laying out in greater detail the policy for accommodating NCB. - Adjusting the definition of concessionality. - For countries that do not normally rely on concessional financing: - No change to tailored approach, but improved debt transparency would allow for more rigorous identification of vulnerabilities. - Core principle retained: - Debt conditionality remains linked to DSAs, risk ratings, and identified vulnerabilities. - For countries that normally rely on concessional financing: conditionality not expected in low risk of debt distress cases but otherwise required. - For countries that do not normally rely on concessional financing: conditionality expected when needed to address critical debt vulnerabilities. - Debt limits to continue being set on a contracting or disbursement basis when warranted. --- ### A. Debt transparency — enhancing debt data disclosure to staff (A.1) - Revised policy expectation: - Critical debt data disclosure gaps should be addressed upfront in IMF-supported programs. - In line with Guidelines on Conditionality, prior actions could be appropriate. - Additional debt-related information may be obtained over the course of the program using structural benchmarks where warranted. - Technical assistance to support authorities’ efforts to enhance debt data disclosure where appropriate. - Risk-based approach with “red flags” to prioritize disclosure efforts: - Indications of significant omitted debt: - (i) state and/or local governments not already captured in debt numbers that have capacity to borrow (and indications they have used this capacity); - (ii) large SOEs with capacity to borrow externally not already captured in the debt perimeter (and indications they have used this capacity); - (iii) PPPs or SPVs not adequately evaluated or monitored. - Missing information about terms and conditions of significant debt (or significant contingent fiscal liabilities, such as PPPs), combined with indications that vulnerability-inducing or vulnerability-enhancing conditions could exist (e.g., collateral, margin call clauses, step-up rates, non-standard event-of-default clauses). - Conditionality from disclosure: - Debt disclosure would lead to conditionality only if the vulnerability is deemed critical for IMF-supported program implementation; requires case-by-case approach. - Not all public sector entities would be subject to conditionality—commercially-viable SOEs continue to be excluded (see Box 4). --- ### A.2. Enhancing transparency on debt holders’ profile - Rationale: - Creditor composition information helps assess reliance on international financial markets, domestic and foreign banks, bilateral, plurilateral, and multilateral creditors. - Needed to better assess rollover risks and nature of debt vulnerabilities and to calibrate debt conditionality. - Useful for understanding burden-sharing patterns across additional financing sources within an IMF-supported program. - Policy change: - Revised policy would require program documents for all IMF-supported arrangements to include a table providing a profile of holders of the country’s public debt. - Annex I (in the original document) provides an example of the proposed content of the debt holders’ table. - Table should include the most recent data and be updated at each program review, subject to confidentiality in underlying debt contracts. - Collateralized debt to be included as a memo item due to separate risks. - Missing elements expected to be filled in during early program stages (by second review at latest); technical assistance could be mobilized if needed. - Expected benefits: - Research indicates increasing transparency standards causally reduces borrowing costs and improves debt management. - Evidence cited: subscription to IMF’s data standards estimated to reduce sovereign spreads by 13 percent over one year. - Increased transparency can speed restructuring/default resolution and lower associated output losses. --- ### Box 4 summary — Debt limits and commercially viable public sector entities - Commercially viable SOEs and similar entities typically excluded from debt limits. - When SOE borrowing activity is modest and fiscal risk to general government is limited, exclusion from debt limits may be preferable to burdensome conditionality. - Case for selective exclusion for commercially viable SOEs that: - May borrow without government guarantee and whose operations pose limited fiscal risk (similar to LIC DSF treatment). - Assessment criteria for exclusion include: - (i) whether SOE carries out uncompensated quasi-fiscal activities; - (ii) whether SOE has negative operating balances. - Additional indicators: managerial independence, relations with government (e.g., offering unrelated collateral), audit periodicity, publication of comprehensive annual reports, protection of shareholders’ rights, financial indices and sustainability, other risk factors. --- ### B. Reforms covering countries that normally rely on concessional financing #### B.1. Countries with significant access to international financial markets — conditionality approach - For countries at high and moderate risk of debt distress: - Default approach: use a PC specified in PV terms on the accumulation of public and publicly guaranteed external debt. - The PV limit is derived from a public borrowing plan, integral to the country’s fiscal framework and consistent with maintaining debt sustainability over the medium-term. - Borrowing space available depends on extent and nature of debt vulnerabilities identified in the DSA. - Alternative formulations of debt conditionality may be used where warranted; staff must justify alternatives using the DSA and judgment. - Example: a QPC on the aggregate level of public debt specified in nominal terms where there is substantial foreign investor participation in the domestic bond market (already allowed under the DLP for countries with an open capital account and significant financial integration with international markets). - Illustrative Guide to Tailoring Conditionality (supplement) provides examples for tailoring; determinations remain case-by-case. #### Eligibility for “significant access to international financial markets” - Stability and risk considerations suggest a bar be met to declare a country has “access to international financial markets on a significant scale.” - Eligibility conditions (both required; country status can shift over time, even within a program): - (i) The country has had (a) significant access to international financial markets in recent years or (b) access to these markets is a key element of the program: - (a) “Significant access to international financial markets” would entail cumulative borrowing of at least 50 percent of quota in at least two of the preceding three years. --- *Italic: IMF staff report content as provided in the PDF chapter/section.* ### Section Illustrative Characterization of Countries Relying on Concessional ### Section Illustrative Characterization of Countries Relying on Concessional Financing with Significant Access to International Financial Markets ### Definition of significant access and operational criteria - Access definition: access to international financial markets would meet the criterion if the program envisages financing on international financial markets in the amount of at least 100 percent of quota over the course of the program period. - The 100 percent of quota threshold applies only where the country does not meet the criterion of 50 percent of quota in prior borrowing. - Alternate condition: the access definition is also met if access to these markets is an integral component of the member’s borrowing plan or of the baseline projections of the Medium-Term Debt Strategy (MTDS). - Demonstrated capacity requirement: the member has demonstrated the capacity to manage significant levels of market borrowing, supported for example by the existence of an MTDS and an actively updated annual borrowing plan. ### Expected benefits of tailoring for countries with significant market access - Better alignment of program conditionality with the country’s financing mix and program design. - Greater ease in conducting refinancing operations. - Incentivizing countries to improve debt management capacity (to obtain such treatment). - Better bridge to debt limits treatment of countries that do not normally rely on concessional financing (where tailoring is standard). ### Countries without significant access to international financial markets — moderate risk of debt distress: expand use of PV limits - General expectation: capacity of countries to set conditionality on aggregate debt levels (including in PV terms) is adequate; broader use of PV limits consistent with the 2014 DLP reform. - If capacity is assessed as not adequate, retain NCB-based policy regime: - Assessment based on debt management capacity evidence (World Bank DEMPAs, IMF technical assistance reports, UNCTAD DMFAS, COMSEC reports, IMF staff experience). - Assessment basis must be clearly explained in staff report. - In such cases: a PC specified in nominal terms set on NCB, and a memo item specified in nominal terms set on CB. - Higher scrutiny required where countries at moderate risk have limited space to absorb shocks: - Borrowing plans remain required and should include a buffer to accommodate shocks without tripping a country into high risk. - Buffer options: adequate flexibility in project execution or appropriate state-contingent features in debt. - Expected benefits: - Expand flexibility provided by PV limits to a wider set of countries. - Provide incentives to optimize borrowing terms (greater volumes if loans have higher concessionality). - Eliminate distortive threshold effects (current rule treats financing with concessionality even slightly below 35 percent as non-concessional). - Enhanced monitoring for countries with limited shock-absorption space as a safeguard. ### Countries without significant access — high risk of debt distress: clarity on NCB exceptions - Presumption: retain zero limit on NCB for countries without significant access that are assessed at high risk of debt distress. - Focus: encourage concessional financing; NCB is less common and typically project-oriented. - Proposed greater clarity on circumstances for NCB exceptions: - Projects: exceptions focused on projects integral to authorities’ national development program—credibly projected to yield a high social and economic return—for which concessional financing is not available. Staff proposes a signal-based approach to determine project exceptions. - Sectoral project assessments by MDBs or credible independent evaluators can inform when a project is integral to the development program, but are not the only allowable signal. - Debt management operations: require demonstration via the DSA that the operation improves key liquidity and/or solvency debt burden indicators without adversely affecting the risk rating. - Repeated NCB exception requests: further assessment required; evidence of past poor use of exceptions raises the bar (additional signal required). Repeated debt management exceptions may trigger assessment of whether country should be reclassified as accessing international financial markets on a significant scale (with default non-zero NCB-limits and possibility to tailor further). - Safeguard: require an indicative target (IT) on public external borrowing specified in PV terms; DSA used to inform PV limit. In case of capacity constraints, existing memo item on CB retained. - Expected benefits: - Strengthen IMF ability to balance whether NCB exceptions should be granted and in what volume. - Mitigate potential for uneven treatment across countries by establishing objective signals. - Make process more transparent and consistent for authorities, Executive Board, and staff. ### Box 5 — Signal-based approach for NCB project exceptions (indicative core signals) - Principle: require at least one signal indicating project is integral to the development program and one signal indicating concessional financing is not available. - Indicative signals for determining what projects are critical for national development: - The project is commercial in nature (e.g., a joint venture) such that commercial revenues would offset debt service costs. - The project is part of an existing Public Sector Investment Program (PSIP) or National Development Plan (NDP). - The project is listed in country's Disaster Resilience Strategy (DRS). - The project is part of the authorities Covid-19 response or mitigation strategy. - The project is endorsed/undertaken by an MDB. - The country is assessed as an adequate performer (above average institutional and efficiency scores for LICs, >=2) under a recent Public Investment Management Assessment (PIMA). - A credible independent assessment is available showing good social and economic returns. - Indicative signals for establishing no alternative concessional financing: - A recent donor conference. - Recent policy/program consultations with key development partners that provide concessional funding (World Bank, regional development banks, bilateral development agencies). - Recent transmission of documentation from the country authorities to MDB(s) on development priorities. - An explicit indication that MDB(s) cannot provide concessional financing to the country in question. - Process notes: - Quality of any signal must be assessed case-by-case during internal review and program documents must state why signal(s) were informative. - If a signal’s quality is questioned (e.g., NDP not up to date/credible), require a second signal. - Guidance Note for staff will describe assessment of these signals in detail. ### Discretion to introduce additional conditionality - The revised policy for countries that normally rely on concessional external financing preserves option to add conditionality to address specific debt vulnerabilities. - Additional conditionality could address domestic debt vulnerabilities, collateral, specific contingent liabilities (e.g., from SOEs/PPPs). - Any additional conditionality must address critical vulnerabilities and be subject to parsimony requirements under the Guidelines of Conditionality. - Section Illustrative Guide to Tailoring Conditionality of the supplement provides illustrative examples. ### Adjusting the definition of concessionality — three proposed changes - Blended financing with financially significant grants in kind: - Such arrangements would be treated as non-concessional (with zero grant element). - Rationale: without international competitive bidding among multiple competitors, fair value of grants in kind (e.g., equipment, machines, engineering services) is difficult to assess; neutral third-party evaluations not operationally feasible. - Note: proposal relates only to cases where grants in kind form part of blended financing packages. - Financing involving collateral: - Some uncommon financing involving collateral would be treated as non-concessional (with zero grant element), specifically when collateral is unrelated to the transaction (e.g., general budgetary borrowing collateralized by earmarking of commodity receipts). - Distinction: ‘related’ collateral = asset financed or future flow receivable/revenue stream generated by the project; ‘unrelated’ collateral reduces repayment risk by collateral outside project. - Measuring concessionality of unrelated collateralized debt requires factoring elements not easily valued/costed; proposed treatment is to treat such loans as non-concessional at this juncture to address potential circumvention and debt vulnerability buildup. - Note: only concessionality treatment and present value measurement affected; no restrictions on use of such transactions implied. - Elimination of the “high concessionality tool”: - Under the proposal, the standard concessionality threshold would apply to all cases: 35 percent. - Rationale: single definition simplifies framework; little loss in ability to catalyze concessional financing. - A higher concessionality threshold would still be allowed when deemed integral to restoring debt sustainability. ### Countries that do not normally rely on concessional financing - Policy remains unchanged for how debt conditionality is set for countries that use the MAC DSA. - In many such countries, fiscal conditionality will continue to suffice if there is comprehensive fiscal coverage and sufficient debt management capacity to handle sophisticated debt structures. - When debt conditionality is critical for the IMF-supported program and not adequately addressed by fiscal conditionality, tailored debt conditionality to specific debt vulnerabilities will be used. - Reforms to encourage greater debt disclosure provide a basis to more rigorously adapt debt conditionality to country-specific circumstances. ### Debt management considerations and incentives - Incentives to improve debt management capacity: - Access to flexible PV limits remains conditioned on adequate capacity to record and monitor debt. - Operating within PV limits incentivizes improving broader debt management functions to maximize borrowing space. - Requirements for debt disclosure support improvements in legal frameworks and procedures for monitoring, recording, and reporting off-balance-sheet exposures. - For countries with significant market access, MTDS adoption and active annual borrowing plan would be requirements to access more flexible tailored conditionality. - Where significant weaknesses in debt management capacity are identified: - Structural conditionality to address deficiencies is expected to feature in associated IMF-supported programs, consistent with Guidelines on Conditionality. - Coordination opportunities with the World Bank and other providers for capacity development support (including IMF technical assistance) exist; World Bank and multilateral agencies (UNCTAD, COMSEC) have specialized capacity for debt management development. - When capacity development is needed: - Support can accompany IMF-supported programs or be accessed through other providers where available. *Source: ppea2020061 - Section Illustrative Characterization of Countries Relying on Concessional Financing with Significant Access to International Financial Markets (supplement).* ### 59. Under the reform proposals discussed, the architecture of the DLP would evolve rather ### REFORM OF THE POLICY ON PUBLIC DEBT LIMITS IN IMF-SUPPORTED PROGRAMS ### Overview of proposed reforms - The architecture of the DLP would evolve rather than change radically; successful existing practices would be standardized. - IMF policy already: - Encourages a focus on debt data disclosure issues. - Allows for the use of PV limits and accommodates non-zero NCB limits where exceptions can be justified. - Provides limited tailoring for countries integrated into international financial markets. - The thrust of reforms is to ensure these practices become standard. ### High-level balance and objectives - Viewed as a package, the proposed reforms are described as balanced: providing more flexibility to manage debt while installing safeguards to support debt sustainability. - Key elements highlighted: - More flexibility where appropriate (e.g., tailored use of PV limits, NCB exceptions for critical projects or debt-management operations). - Safeguards to avoid excessive debt accumulation (e.g., higher scrutiny where borrowing plans place a country in “limited space to absorb shocks” categories, tighter rules on repeated NCB exceptions). - Greater debt data disclosure to the IMF and tighter concessionality definition across all countries. ### COVID-19 shock: implications and policy fit - Anticipated effects of the global COVID-19 pandemic shock: - A change in the distribution of countries across risk-of-debt-distress categories, with more countries moving into high risk of debt distress or in debt distress; some in debt distress may return to at least moderate risk after restructurings. - A likely resumption of the trend away from concessional financing as markets re-open and donor capacity is constrained, increasing the importance of handling high-risk market-access borrowers and encouraging stronger debt management practices. - Materialization of contingent liabilities, underscoring the need to strengthen debt transparency—a fundamental pillar of the reforms. - Table 2 summary (as presented in the source) outlines: - For moderate-risk countries: PC on PV limits of external borrowing used in more countries to incentivize concessional financing and eliminate distortive threshold effects; higher scrutiny where borrowing plans worsen shock-absorption capacity. - For high-risk countries: Clear framework for NCB exceptions where access to international financial markets is limited; PC on PV limits in market-integrated countries; IT on PV limits where countries lack significant market access to prevent exceptions from leading to excessive debt; safeguards on repeated NCB exceptions; conditions for tailored conditionality in market-integrated countries. - For all countries: greater debt data disclosure to the IMF and tighter concessionality definition. - Note: “A borrowing plan should not move a country to high risk as at present.” ### Alignment with World Bank Sustainable Development Finance Policy (SDFP) - The revised DLP and the World Bank’s SDFP would continue to be closely aligned. - Coordination points: - Staff from both institutions would coordinate to align debt policy commitments in IMF-supported programs with SDFP commitments. - When initiating a new IMF-supported program, program design would take into account borrowing limits set under the SDFP together with an updated assessment of debt vulnerabilities. - The SDFP would, in principle, align borrowing limits with the DLP when there is an IMF-supported program and would take account of DLP limits when an IMF program expires. - Exceptions for NCB would be broadly aligned: both policies would only allow NCB exceptions related to critical projects where concessional financing is not available, and/or related to debt management operations. - Complementarity: IMF programs are positioned to obtain debt information systematically; World Bank operations are often better placed to address legal and institutional capacity shortfalls causing inadequate transparency. ### Transitional arrangements and timing - Policy effective date mechanics: - Proposed that the policy take effect at the end of the month following issuance of an updated staff Guidance Note. - It is expected the Guidance Note would be issued in February 2021; therefore the policy would take effect in March 2021. - Conditionality in pre-existing IMF-supported programs: - Conditionality on the basis of new guidelines would be modified in program discussions concluded after March 31, 2021, once modifications have been agreed between staff and authorities and approved by the Executive Board. - Programs approaching their penultimate or last reviews would not be required to change established framing of conditionality. - Staff will continue outreach on main elements of the proposed policy, including after rollout. - Review clause: - A review of implementation experience would be conducted no later than five years after the entrance into effect of the new policy. ### Implementation practice and experience (selected findings) - Application of the DLP generally followed DSA risk ratings or signals: - PRGT cases: - High-risk countries: DLP applied as intended with zero NCB limits; where debt recording/monitoring capacity adequate, additional limits on concessional borrowing (PCs or ITs) were used. - Moderate-risk countries: policy applied more restrictively than envisaged in some cases; not all countries featured PV limits owing to weak debt recording/monitoring capacity; several had zero-NCB nominal limits. - Low-risk cases: generally no DLs except targeted limits on SOE debt or government guarantees in some programs. - GRA cases: - Programs with significant vulnerabilities generally featured DLs tailored to the nature of vulnerability (e.g., ceilings on total public debt, external debt, guaranteed or SOE debt, short-term or collateralized debt). - A few cases that lacked DLs may have benefited from them (examples cited include Seychelles 2017 and Tunisia). - Examples of policy flexibility in practice: programs have used exceptions and adjusters on NCB, sometimes added during reviews (e.g., Afghanistan, Cameroon, Ghana, Mauritania). ### Issues for Board discussion (as listed in source) - Do Directors agree with proposals to enhance debt disclosure to improve the design of IMF-supported programs, including specification of debt limits? - Do Directors agree that the proposed policy for countries that normally rely on concessional financing but have recently started accessing international financial markets on a significant scale would better align program conditionality with the country’s financing mix and program design? - Do Directors agree with the proposal to revise eligibility criteria for the use of PV limits? - Do Directors agree that a signal-based approach should be used for NCB exceptions in high-risk countries to make the process more transparent and consistent? - Do Directors agree that proposals to modify the measurement of “concessionality” are warranted, including to help prevent circumvention of debt limits? *Source: Reform of the Policy on Public Debt Limits in IMF-Supported Programs (excerpt).* ### 2. Observance of DLs since the implementation of the 2014 DLP has generally been ### ppea2020061 - 2. Observance of DLs since the implementation of the 2014 DLP has generally been ### Observance of debt limits and breaches (July 2015–April 2020) - Debt PCs between July 2015 and April 2020 were breached in 8 programs with at least one review, representing approximately 4 percent of all test dates. - With the exception of Malawi, the size of the breaches were moderate relative to GDP, but in some cases they were large relative to the debt limits (particularly in Cameroon). - Main reasons for non-observance of debt limits: - Weak public financial management and debt monitoring (Cameroon, 2017 (two occasions); Iraq, 2016). - Reporting slippages (Guinea, 2017; Liberia, 2015; Mozambique, 2015). - Technical factors (Malawi, 2018; Mauritania, 2017). - Of the four programs that went off-track: - Haiti 2015, Mozambique 2015, and Sierra Leone 2017 went off-track before the first review. - Kenya 2015 went off-track after the second review. - One case was due to non-observance of the program debt limit (Mozambique, 2015); the others were due to exogenous shocks or weak fiscal performance. - Compliance with debt-related ITs has also been strong, with only three instances of missed targets during July 2015–April 2020 (Egypt, Rwanda, and Sri Lanka). ### Selected breaches from Table 4 (size, reason, corrective actions) - Guinea — Nominal non-zero NCB, Review R2, Size: 1 percent of limit / 0.1 percent of GDP - Reason: Investment — Signed previously undiscussed loan for key infrastructure project; loan not submitted for ratification, negotiations reopened. - Corrective action: Staff modified the QPC to exclude debt that is non-concessional upon signature but later cancelled or renegotiated to become concessional. - Cameroon — Nominal non-zero NCB, Review R1, Size: 64 percent of limit / 0.3 percent of GDP - Reason: Investment — Authorities advanced loan agreement initially included in the borrowing plan for the following year (infrastructure for Africa Cup). - Corrective action: Limit for the following year was lowered by the same amount. - Cameroon — Disbursements of NCB, Review R2, Size: 16 percent of limit / 0.4 percent of GDP - Reason: Technical — Actual disbursements on major development projects were higher than anticipated (communications problems with some financial partners). - Corrective action: Measures to better anticipate recording of these disbursements through better planning and monitoring. - Iraq — Gross public debt, Review R2, Size: 4 percent of limit / 0.3 percent of GDP - Reason: Budget/Investment — Budget deficit higher than programmed and authorities issued guarantees for infrastructure projects. - Corrective action: Prior actions to improve fiscal discipline. - Liberia — PV Limit, Review R7, Size: 88 percent of limit / 0.4 percent of GDP - Reason: Reporting — Reporting slippage (project loan ratified by the legislature not included in the total amount). - Corrective action: Debt management unit began monthly reporting on all signed/ratified loans. - Malawi — Zero NCB Limit, Review R1, Size: US$127 million / 1.9 percent of GDP - Reason: Technical — Technical oversight in the TMU when nonresident bank purchased domestically denominated Treasury Notes that constituted NCB. - Corrective action: TMU modified going forward to exclude domestically denominated Treasury Notes and Bills from the QPC. - Mauritania — Zero NCB Limit (w/exception), Review R4, Size: 0.02 percent of limit / <0.1 percent of GDP - Reason: Technical — Actual value of NCB loan exceeded allowed amount by 2 percent when valued at program exchange rates. - Corrective action: None (waiver granted on basis of there being no material impact on the DSA). - Mozambique — PV Limit, Review R1, Size: n.a. / n.a. - Reason: Reporting — Disclosure of hidden debt by SOEs revealed breach of NCB limit under previous program. - Corrective action: N.A. program went off track. ### Accommodation and modifications of debt limits - Upward revisions of borrowing limits or adjusters to accommodate infrastructure investment or debt management operations were limited to a small group of countries. - Upward modifications under IMF-supported programs occurred in Afghanistan, Cameroon, Mauritania, and repeatedly in Ghana. - Debt-related conditionality modifications did not lead to deterioration in risk ratings; improvements were not observed either. - On the debt management side: - Côte d’Ivoire and Ghana had multiple modifications to accommodate Eurobond issuances. - Togo’s program allowed for some NCB to pay down more costly domestic debt. - Other modifications related to changes in projections, capacity or DSA upgrades, technical factors, and missed PCs. ### Stakeholder survey and consultation results — roundtables and surveys - Roundtable discussions in 2019 highlighted: - Concerns about inadequate concessional resources; need to explain staff proposals in that context. - Support for enhancing debt data transparency and broader use of PV limits. - Advocacy for taking on more risk in moderate-risk countries to promote investment and growth, while noting that zero NCB limits may remain appropriate in high-risk countries with concessional financing access. - Recognition that NCB exceptions can be important in some high-risk countries; suggestion to strengthen capacity for project selection and implementation (e.g., developing PIMA-like frameworks). - Broad support for better tailoring debt limits to country vulnerabilities. - Annual Meetings roundtable with country authorities: - Feedback generally positive; delegates emphasized that the DLP helped prioritize investments, led to more careful borrowing-plan review, increased focus on favorable borrowing terms, and improved central government control over SOE debt contracting. - Delegates underscored the importance of fiscal targets, sound financial management institutions, strong governance, debt management, and transparency. - Delegates called for better understanding of country-specific challenges and more flexibility in setting and modifying debt limits. - Annual Meetings roundtable with Multilateral Development Banks (MDBs): - MDBs noted rising debt vulnerabilities pose operational challenges. - Limitations of project-by-project evaluation were highlighted, especially where political considerations undermine expected revenues. - Some MDBs advocated more flexibility in DLs for profitable non-concessionally financed projects, particularly where concessional financing is limited, while noting that more flexibility could reduce incentives to provide concessional financing. - Mission chief (MC) survey (December 2018): - 33 responses were received. - Most respondents considered the DLP effective or somewhat effective in containing debt vulnerabilities by: (i) limiting NCB; (ii) limiting borrowing in general; and (iii) pushing back line ministries. - MCs considered that the DLP complemented fiscal conditionality by capturing: (i) contracting of debt; (ii) guarantees; and (iii) specific borrowing terms (short-term debt, non-concessional debt, collateralized debt). - MCs critical of the DLP noted: (i) it was overdetermined and fiscal conditionality would have sufficed; and (ii) it did not prevent authorities from concealing debt. - Issues identified needing further refinement: - The policy constrains investment (including by limiting semi-concessional loans). - It creates incentives to avoid requesting a program and borrow from non-concessional external and domestic sources. - It leads to nontransparent practices to circumvent conditionality (concerns with PPPs and collateralized borrowing). - Implementation challenges point to need for further guidance, government practice issues, and policy design issues. - Less than 10 percent of MCs indicated the DLP was effective in encouraging progress in strengthening debt recording and monitoring capacity; 45 percent indicated it was somewhat effective. - Reasons for limited effectiveness in strengthening recording/monitoring included other priorities, time required for capacity building, and that the IMF does not provide TA in this area. - Surveys of country authorities and creditors: - Country authority survey: only two respondents replied. - Creditor survey: of 53 creditors contacted, 7 with significant lending activities responded. - 71 percent of creditor respondents indicated they had a strong understanding of how the DLP applies to their operations, the role of the authorities, and how to contact the IMF/WB for clarification. - Views on DLP effectiveness in containing debt vulnerabilities were mixed: effective (33 percent), somewhat effective (50 percent), not effective (17 percent). - Indications that respondents adapted financing to comply with the DLP were limited. - Most (71 percent) indicated they did not have enough information to check compliance with debt limits conditionality agreed between authorities and the IMF. ### Illustrative guide to tailoring conditionality - Circumstances where tailored debt conditionality could be invoked: - As the standard approach for countries that do not normally rely on concessional financing. - As a potential approach for countries that normally rely on concessional financing but regularly tap international financial markets. - As an optional approach to target specific debt vulnerabilities (not addressed through standard debt or other program conditionality) for countries that normally rely on concessional financing. - Tailoring is case-by-case and underpinned by analysis of debt-related risks: - DSAs distinguish three risk types: (i) risks from elevated debt levels; (ii) risks related to gross financing needs (GFNs); and (iii) risks from debt structure (maturity, guarantees and contingent liabilities, currency composition, collateralization). - Risks can be concentrated in domestic or external debt. - MAC-DSA and LIC-DSF produce indicators informative on vulnerabilities along these dimensions. - Illustrative forms of debt conditionality by area of vulnerability (examples): - High debt level - Domestic: Limit on domestic borrowing to substitute with external borrowing and/or encourage refinancing at longer maturities. - External: Limit on external borrowing to substitute with domestic borrowing. - Domestic + external: Fiscal conditionality augmented with a limit on external borrowing; encourage refinancing at longer maturities. - High GFN - Domestic: (Most naturally addressed through fiscal conditionality) Limit on domestic borrowing; encourage refinancing at longer maturities. - External: Limit on external borrowing; encourage refinancing at longer maturities. - Debt structure - Maturity: Limit on short-term borrowing; limit on issuing debt maturing in specific years; target for future average maturity of debt stock. - Contingent liabilities: Limit on government guarantees; limit on borrowing by vulnerable SOEs; limit on size of PPP programs; broaden fiscal indicator coverage to encompass the source of vulnerability. - Currency composition: Limit on foreign-currency denominated borrowing. - Collateralized debt: Disclosure requirements; limit or prohibit new issuances of collateralized debt; set target to reduce existing stock of collateralized debt. _RefORM OF THE POLICY ON PUBLIC DEBT LIMITS IN IMF-SUPPORTED PROGRAMS—SUPPLEMENTARY INFORMATION, INTERNATIONAL MONETARY FUND_ ### 11. The following describes the criteria that are proposed for use in identifying countries ### ppea2020061 - 11. The following describes the criteria that are proposed for use in identifying countries eligible for the proposed tailoring approach ### Eligibility criteria for “significant access to international financial markets” - Proposed approach draws on measures used for determining market access for PRGT blending and graduation purposes and aims for consistency with IMF access policies and World Bank IDA eligibility. - Definition option 1: - Cumulative borrowing of at least 50 percent of quota in at least two of the preceding three years. - Definition option 2 (alternative where prior borrowing criterion is not met): - Access to international financial markets is an integral component of the member’s borrowing plan or of the baseline projections of the Medium-Term Debt Strategy (MTDS), and - The program envisages financing on international financial markets in the amount of at least 100 percent of quota over the course of the program period. - This 100 percent of quota threshold applies only when the 50 percent of quota in prior borrowing criterion is not satisfied. - The eligibility is not fixed and would vary depending on recent developments and the financing strategy underlying a program. ### Illustrative statistics and factual findings - Of the 70 PRGT-eligible countries there are 14 that currently meet the proposed significant market access definition based on past data (Table 8). - Based on 2016–18 data, NCB had an average share of 37 percent of total new borrowing in these countries. - The analysis uses the World Bank’s International Debt Statistics (IDS) database as the default (noting a typical ten-month lag). - Most of the identified countries are also presumed or potential PRGT blenders except for Benin and Ethiopia which do not meet the additional income criteria. ### World Bank Sustainable Development Finance Policy (SDFP) — overview - In June 2020, the World Bank approved transitioning from the Non-Concessional Borrowing Policy (NCBP) to the Sustainable Development Finance Policy (SDFP). - Objectives of the SDFP: - Support IDA clients to strengthen policies, institutions, and practices for transparent and sustainable financing of development goals. - Enhance coordination among borrowers, creditors and other development partners. - Introduce a more robust monitoring and accountability framework. - Country coverage expanded to all IDA-eligible countries (post-MDRI and IDA grant recipients no longer the only group covered). - SDFP centers on two pillars: - Debt Sustainability Enhancement Program (DSEP). - Program of Creditor Outreach (PCO). ### DSEP pillar (SDFP) — key features and near-term focus - DSEP defines annual performance and policy actions (PPAs) to strengthen debt transparency, fiscal sustainability, and debt management for all IDA countries at moderate or high risk of debt distress (calibrated to debt vulnerabilities and capacity constraints). - Near-term focus (post-COVID-19 environment): improving debt data transparency and debt management. - Implementation mechanics: - If progress on PPAs is unsatisfactory after the first year, a portion of the IDA allocation will be set aside. - If progress is satisfactory after the second year, the allocation set aside during the first year will be released. - Timing: assessment of PPAs concluded at end of FY2021 and set asides kick off at beginning of FY2022. - Exclusions from PPA and set-aside system include: low risk of debt distress under LIC-DSF; countries under the MAC DSA where World Bank Management determines limited vulnerabilities; countries without regular IDA programs due to loans/credits in non-accrual status; countries eligible for IDA’s Remaining Engaged in Conflict Allocation. - Set-aside amounts: for moderate and high-risk countries the set aside would be 10 and 20 percent of IDA allocations, respectively. For blend and gap countries under the MAC DSA there will be a set aside of 10 percent unless WB management determines that the country’s debt vulnerabilities are limited. ### PCO pillar (SDFP) — key features - Promotes stronger collective action among borrowers, creditors, and international development partners. - Specific actions include: - Promotion of “Core Principles of Sustainable Financing” across MDBs and IFIs. - Expanding MDB outreach boundaries. - Strengthening outreach and communication with bilateral creditors on the SDFP. - Promoting dialogue on sustainable financing and debt transparency with Paris Club, non-Paris Club and private creditors. - Strengthening dialogue and coordination around debt-related TA. - Enhancing use of the Lending to LICs platform. ### SDFP limits and exceptions - Debt limits would primarily take the form of nominal limits on non-concessional external PPG for countries at moderate or high risk of debt distress. - PV limits on external borrowing could be considered if circumstances warrant it. - Low risk countries would normally not be subject to a ceiling, but a ceiling could be triggered under certain circumstances (e.g., a rapid debt build-up). - Ex-ante exceptions may be granted where borrowing is linked to projects with high economic and social returns; require comprehensive information on financing and project impact (volume, financing terms, disbursement schedule, use of proceeds, available project analysis). - Ex-post exceptions would not be granted in principle, unless financing was in response to an exogenous shock or the project meets ex-ante exception criteria and a rationale is provided for why it was not discussed ex-ante. ### Complementarity between World Bank SDFP and IMF DLP (key alignment points) - Staff coordination to ensure close alignment between debt policy commitments in IMF-supported programs and those specified in the SDFP. - When initiating a new IMF-supported program, program design would take account of borrowing limits set under the SDFP together with an updated assessment of debt vulnerabilities. - SDFP maintains approach of aligning, in principle, borrowing limits with the DLP when there is an IMF-supported program; when an IMF-supported program expires, SDFP would take account of borrowing limits set under the DLP together with an updated assessment of debt vulnerabilities. - Exceptions for NCB broadly aligned: both DLP and SDFP would only allow NCB exceptions related to critical projects where concessional financing is not available, and/or related to debt management operations. - Policies complement each other in promoting improved debt transparency, fiscal sustainability, debt management, and governance: - IMF-supported programs often obtain relevant debt information systematically. - World Bank operations often better positioned to address legal and institutional problems behind inadequate transparency and capacity shortfalls. - Calibration of limits for countries normally relying on concessional financing under both policies would be underpinned by the joint LIC-DSF. - For countries not normally relying on concessional financing, calibration of limits (if any) would be supported by the IMF’s MAC DSA for the DLP and the World Bank’s analysis of debt vulnerabilities for the SDFP. ### Comparative highlights (SDFP vs IMF DLP under proposed reforms) - Duration: - SDFP: Applies continuously. - IMF DLP w/Proposed Reforms: Applies exclusively under IMF-supported program. - Country coverage: - SDFP: All countries eligible for IDA resources (74 countries during IDA19). - IMF DLP: IMF membership (189 countries). - Available instruments: - SDFP: Set-aside portion of IDA-allocation to incentivize completion of PPAs, which include debt limits (consistent with IMF DLP). Hardening of terms may apply in cases of severe or repeated breaches. - IMF DLP: Under IMF program: performance criterion on non-concessional borrowing or present value of debt. - Framework for debt ceilings by LIC-DSF category (selected points): - LIC-DSF low risk: - SDFP: Normally not subject to ceiling unless warranted. - IMF DLP: Ceiling not required (option for targeted if needed). - LIC-DSF moderate risk: - SDFP: Normally nominal NCB ceiling to avoid risk of external debt distress downgrade. - IMF DLP: PC on PV limit of external borrowing (in most cases) calibrated to avoid external debt distress downgrade. - LIC-DSF high risk (sustainable outlook): - SDFP: A zero ceiling on NCB borrowing will apply in principle; NCB allowed only under exceptional circumstances (e.g., debt management operations). - IMF DLP: Zero NCB for countries without significant access to international financial markets (with NCB exceptions allowed under limited circumstances). PV on external borrowing for countries with significant access to international financial markets (or alternative if better targeted). - LIC-DSF in debt distress or at high risk (unsustainable outlook): - SDFP: A zero ceiling on NCB applies. A non-zero ceiling could be considered (e.g., for arrears clearance operations only). - IMF DLP: N.A. (IMF lending prohibited). - NCB exceptions: - SDFP: Ex-ante exceptions for projects with high economic and social returns when concessional financing is not available and if operation is needed for debt management operations that do not lead to a worsening of the debt profile, and consistent with the MTDS and MTFP (if exists). Ex-post exceptions only if in response to exogenous shock or if project meets ex-ante criteria and rationale for not being discussed ex-ante. - IMF DLP: Exceptions for projects integral to the authorities' development program where concessional financing is not available (assessed using a signals-based approach) or debt management operations that result in the overall improvement of the debt profile. - Institutional process: - SDFP: Management-driven given continuous application (Board regularly informed to promote oversight). - IMF DLP: Board-driven as debt limits are an integral part of IMF programs. *Source: Excerpt from ppea2020061 — Reform of the Policy on Public Debt Limits in IMF-Supported Programs — Supplementary Information (text provided).* --- _Source: https://www.imf.org/-/media/files/publications/pp/2020/english/ppea2020061.pdf_